exceptions_exp2_swap_0.3_resemble_to_hit_2128

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5633
• Accuracy: 0.3687

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0006
• train_batch_size: 16
• eval_batch_size: 16
• seed: 2128
• gradient_accumulation_steps: 5
• total_train_batch_size: 80
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 50.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
4.815	0.2915	1000	0.2535	4.7594
4.3399	0.5830	2000	0.2990	4.2841
4.152	0.8745	3000	0.3144	4.1015
3.9962	1.1659	4000	0.3243	3.9958
3.9504	1.4574	5000	0.3304	3.9211
3.8819	1.7488	6000	0.3360	3.8620
3.7462	2.0402	7000	0.3405	3.8210
3.7613	2.3317	8000	0.3436	3.7897
3.7417	2.6232	9000	0.3462	3.7610
3.7271	2.9147	10000	0.3484	3.7343
3.6424	3.2061	11000	0.3503	3.7222
3.6548	3.4976	12000	0.3522	3.7038
3.648	3.7891	13000	0.3541	3.6842
3.5508	4.0805	14000	0.3552	3.6775
3.5812	4.3719	15000	0.3561	3.6669
3.569	4.6634	16000	0.3573	3.6520
3.5877	4.9549	17000	0.3587	3.6403
3.4948	5.2463	18000	0.3590	3.6430
3.5325	5.5378	19000	0.3600	3.6321
3.5233	5.8293	20000	0.3612	3.6200
3.4489	6.1207	21000	0.3614	3.6258
3.4843	6.4122	22000	0.3619	3.6172
3.4817	6.7037	23000	0.3628	3.6051
3.5031	6.9952	24000	0.3637	3.5966
3.427	7.2865	25000	0.3635	3.6063
3.4648	7.5780	26000	0.3641	3.5961
3.4604	7.8695	27000	0.3649	3.5884
3.3862	8.1609	28000	0.3646	3.5988
3.4311	8.4524	29000	0.3655	3.5892
3.4337	8.7439	30000	0.3658	3.5829
3.3416	9.0353	31000	0.3660	3.5872
3.3808	9.3268	32000	0.3663	3.5857
3.4145	9.6183	33000	0.3666	3.5778
3.4145	9.9098	34000	0.3675	3.5677
3.338	10.2011	35000	0.3673	3.5806
3.3617	10.4926	36000	0.3676	3.5732
3.3788	10.7841	37000	0.3683	3.5642
3.2902	11.0755	38000	0.3681	3.5740
3.3374	11.3670	39000	0.3679	3.5737
3.3726	11.6585	40000	0.3687	3.5633
3.3734	11.9500	41000	0.3693	3.5566
3.3117	12.2414	42000	0.3686	3.5702
3.35	12.5329	43000	0.3692	3.5649
3.3493	12.8243	44000	0.3697	3.5541
3.2673	13.1157	45000	0.3692	3.5707
3.3048	13.4072	46000	0.3694	3.5630
3.3353	13.6987	47000	0.3700	3.5566
3.3384	13.9902	48000	0.3706	3.5497
3.2768	14.2816	49000	0.3701	3.5615
3.3012	14.5731	50000	0.3704	3.5572
3.3365	14.8646	51000	0.3709	3.5476
3.2517	15.1559	52000	0.3702	3.5644
3.2817	15.4474	53000	0.3707	3.5549
3.3105	15.7389	54000	0.3712	3.5488
3.2064	16.0303	55000	0.3707	3.5604
3.2604	16.3218	56000	0.3708	3.5570
3.2866	16.6133	57000	0.3712	3.5530
3.3056	16.9048	58000	0.3719	3.5424
3.2399	17.1962	59000	0.3709	3.5605
3.2597	17.4877	60000	0.3715	3.5511
3.2816	17.7792	61000	0.3718	3.5459
3.2005	18.0705	62000	0.3714	3.5583
3.2376	18.3620	63000	0.3713	3.5548
3.2578	18.6535	64000	0.3718	3.5464
3.2823	18.9450	65000	0.3722	3.5393
3.209	19.2364	66000	0.3719	3.5554
3.2468	19.5279	67000	0.3720	3.5495
3.2702	19.8194	68000	0.3724	3.5422
3.1871	20.1108	69000	0.3717	3.5616
3.2294	20.4023	70000	0.3719	3.5546
3.2532	20.6938	71000	0.3725	3.5447
3.2632	20.9853	72000	0.3731	3.5370
3.2012	21.2766	73000	0.3722	3.5546
3.2272	21.5681	74000	0.3728	3.5471
3.2546	21.8596	75000	0.3729	3.5400
3.1653	22.1510	76000	0.3725	3.5569
3.2096	22.4425	77000	0.3727	3.5522
3.2353	22.7340	78000	0.3728	3.5429
3.128	23.0254	79000	0.3727	3.5549
3.187	23.3169	80000	0.3724	3.5552
3.2108	23.6083	81000	0.3729	3.5457
3.222	23.8998	82000	0.3733	3.5391
3.1629	24.1912	83000	0.3727	3.5599
3.1893	24.4827	84000	0.3730	3.5529
3.211	24.7742	85000	0.3735	3.5381
3.1287	25.0656	86000	0.3727	3.5582
3.1707	25.3571	87000	0.3731	3.5532
3.1943	25.6486	88000	0.3735	3.5434
3.2172	25.9401	89000	0.3739	3.5374
3.1435	26.2314	90000	0.3729	3.5581
3.1584	26.5229	91000	3.5578	0.3729
3.1813	26.8144	92000	3.5505	0.3732
3.1252	27.1061	93000	3.5587	0.3729
3.1641	27.3976	94000	3.5547	0.3730
3.1761	27.6891	95000	3.5441	0.3735
3.2012	27.9806	96000	3.5393	0.3742
3.1311	28.2720	97000	3.5569	0.3733
3.1668	28.5635	98000	3.5514	0.3736
3.1675	28.8550	99000	3.5454	0.3738
3.1173	29.1463	100000	3.5612	0.3731
3.133	29.4378	101000	3.5535	0.3734
3.1742	29.7293	102000	3.5449	0.3741
3.0852	30.0207	103000	3.5563	0.3735
3.1259	30.3122	104000	3.5530	0.3736
3.1595	30.6037	105000	3.5495	0.3741
3.1501	30.8952	106000	3.5422	0.3745
3.1198	31.1866	107000	3.5602	0.3736
3.1232	31.4781	108000	3.5533	0.3740
3.142	31.7695	109000	3.5438	0.3742
3.0683	32.0609	110000	3.5560	0.3738

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4